+(,121/,1(
Citation: 48 Ohio St. L.J. 399 1987
Content downloaded/printed from
HeinOnline (http://heinonline.org)
Wed Aug 11 12:54:46 2010
-- Your use of this HeinOnline PDF indicates your acceptance
of HeinOnline's Terms and Conditions of the license
agreement available at http://heinonline.org/HOL/License
-- The search text of this PDF is generated from
uncorrected OCR text.
-- To obtain permission to use this article beyond the scope
of your HeinOnline license, please use:
https://www.copyright.com/ccc/basicSearch.do?
&operation=go&searchType=0
&lastSearch=simple&all=on&titleOrStdNo=0048-1572
Making Sense of the Liability Insurance Crisis
KENNETH
S. ABRAHAM*
I. INTRODUCTION
Complex issues are often debated in simple terms. Simple arguments can pose
dramatic questions and draw public attention to the issues involved; but they may also
deprive public debate of the depth necessary to an accurate understanding of the
problem at hand. Debate about the recent crisis in liability insurance cost and
availability has elicited both simple explanations and simple solutions. The simple
explanations cite simple causes: the natural progression of the "underwriting cycle"
resulting from the decline in interest rates that began in the early 1980s; 1 price
gouging by the insurance industry; 2 an explosion in tort liability necessitating
dramatic increases in premium rates and making insurance against certain risks
impossible; 3 the incentives to initiate tort claims created by our system that make
increases in the costs of tort liability and in the liability insurance costs that
4
accompany it inevitable.
Simple solutions also abound: limit price competition through regulatory
imposition of "flex" rating to counteract the underwriting cycle;5 repeal the
McCarran-Ferguson Act's exemption of the insurance industry from federal antitrust
regulation; 6 return to the tort law of three decades ago; 7 control attorney's fees in tort
8
cases; and adopt the English rule that a losing plaintiff must pay defendant's costs.
These explanations for the crisis and the solutions they generate are not only
simple, but simplistic. Standing alone, no single diagnosis can account for more than
a portion of the recent crisis, no single explanation can adduce sufficiently complete
and compelling data for support, and no single, plausible solution has the power to
make the crisis disappear. The causes of the crisis are complex, and there are no easy
methods of solving the problem, returning premium rates to their former levels, or
obtaining complete assurance that similar crises will not occur in the future. In what
follows I set out my own view of the causes and possible cures of the crisis by
analyzing the four major explanations cited above, exploring the implications of this
* Professor of Law, University of Virginia School of Law. A.B. 1967, Indiana University; J.D. 1971, Yale
University. This is an expanded version of a presentation made to the workshop on Tort Reform at the 1987 Annual
Meeting of the Association of American Law Schools.
15
ADVISORY
CO\MSSION ONLBamrr INsuRAgcE, STATE
OFNEw YORK,
INSURING
OURFUTURE
1. See I GOVERNOR'S
(1986).
1 (1986).
2. See NATIONAL
INSURANcE
COssUMER
ORGANIzA'loN, THE LIAIMLnYCRISISIN INSURANCE
3. See generally INSURACE SERvicES OFnCE, INSURER
PoF'rABUrrT: THE FACTS
(1986).
4. I derive this theory from remarks by J. Robert S. Pritchard, Dean of the University of Toronto School of Law,
made in Washington, D.C. on Nov. 8, 1986.
supra note I, at 93-98.
5. See I GOVERNOR'S
ADVISORY
Co. .nssioT oN Ltmmrr INssRANcE,
6. See NATIONAL
INsuRANCE
CoNuER ORoANamnON, supra note 2, at 7.
7. See G. PRIEsr, Tne CURENT INsuirAsc CRISIS
ANDMODERN
ToRT LAw (unpublished draft, 1986).
INFORmA-nON
INSTrIrrE, THE
8. See, e.g., the following pamphlets prepared by defense-oriented groups: INSURANcE
TORT L.w REFoRw Poucy STATE.NiENT
5 (1986); R. MALOTr,
ANERICA'S
Lwsurr CRIsIS 4 (1986); THE BUSIEfS ROUNDTABLE,
LIABiLiTS
EXrLoszON:
CAN WE AFFORD
THECosr? 9-10 (1985).
HeinOnline -- 48 Ohio St. L.J. 399 1987
OHIO STATE LAW JOURNAL
[Vol. 48:399
analysis for the solutions that have been proposed, and indicating the points at which
satisfactory analysis is impeded by the absence of data about critical issues.
II.
THE UNDERWRITING CYCLE
For many decades the profitability of the property/casualty industry and the
supply of its product have been subject to rather regular cycles. During the "soft"
portion of this underwriting cycle, premiums are comparatively low and coverage is
comparatively easy to obtain. During the "hard" portion of the cycle premiums rise,
as the capacity of the industry to supply coverage contracts.
The causes of the cycle are complex. One factor is the supply of capital to the
industry. The industry's underwriting "capacity"-the amount of insurance it can
sell-is both legally and prudentially a multiple of its capital surplus. 9 Surplus can be
increased through profits earned in underwriting-when premiums exceed the costs
of claims and expenses under policies sold for those premiums. Surplus can also be
increased by accumulation of earnings on premiums that are invested prior to the time
when claims must be paid. Finally, capacity may be increased even without
increasing surplus through the purchase of reinsurance-in effect, the purchase of
insurance by an insurer.
Each of these factors seems to have figured in the most recent cycle. First,
toward the end of 1984 European reinsurers reduced the amount of coverage they
were willing to offer the American property/casualty industry. This decision may
have been due partly to the strength of the dollar against European currencies. But it
also seems to have been caused by these reinsurers' growing concern over the
expansion of their exposure by American rules of tort liability.
Second, between about 1978 and 1983, when high interest rates prevailed, it was
possible for insurers to incur sizeable underwriting losses and still show an operating
profit by earning high returns on invested premiums. During this period competition
in the property/casualty industry was stiff, and premium levels remained at the same
level or (in certain cases) declined. Proponents of the underwriting cycle theory
suggest that as interest rates declined after 1983 it was inevitable that premiums
would rise, because investment profit could no longer offset previous levels of
underwriting loss.
This explanation has some force. However, the magnitude of recent increases
has been far greater than one would expect if a drop in interest rates alone were
responsible. If changes in interest rates roughly parallel changes in the inflation rate,
and the components of tort recoveries follow the inflation rate, then one would expect
little impact on premiums from changes in the interest rate. This is because any gain
or loss in prospective investment income would be offset by a prospective loss or gain
in underwriting results. Changes in the interest rate should affect premium rates
significantly only when the interest rate does not change as quickly as the inflation
rate. In fact, due to fears about future inflation, United States interest rates did
9. See INsuRcE INFORMATION
INSTwRrE,
BAstc CoNCEPs oF AccouorNG ANDTAXAnONOFPRoPRirnICAsuALW IN-StwNCE
CoMPANIEs
48
(1984).
HeinOnline -- 48 Ohio St. L.J. 400 1987
19871
MAKING SENSE OF THE LIABILITY CRISIS
not decline as quickly after 1983 as did inflation. A premium increase therefore could
have been expected once the difference between these rates narrowed.
Such an increase, however, would have been comparatively modest. Take the
extreme situation: a long-tail line of liability insurance such as medical malpractice
or products liability, in which the average claim is not paid for six to eight years after
a policy is sold and in which investment income is therefore a large component of the
operating profitability of the line. In this case one would expect, very roughly, that
premiums might have increased a maximum of fifty percent when the gap between
the interest and inflation rates narrowed, because of the loss of an "extra" three to
five percent annual investment profit over six to eight years. Even this hypothesis
supposes that in setting their premium rates early in the 1980s insurers were betting
against the capital market's fear that inflation (including inflation in jury awards)
would return later in the decade.
Any premium increases above this fifty percent estimate even in long-tail lines
cannot be ascribed to the interest-rate effect; and in shorter-tail lines, in which
investment income comprises a smaller portion of operating profit or loss, one would
expect the effect to have been even smaller. The withdrawal of reinsurance capital
therefore seems a much more probable cause of the market's entry into a hard portion
of the cycle late in 1984 than does the decline in interest rates. Even both factors
together, however, seem unlikely to account for the severity of the problem that
occurred thereafter.
I.THE
CONsPIRAcY THEORY
The conspiracy theory rejects more innocent explanations for the crisis, and
attributes the large increases in the price of liability insurance that struck in 1985 and
1986 as evidence of price gouging by the property/casualty insurance industry.10
Because the industry is largely exempted from the federal antitrust laws, the theory
suggests that price competition can be avoided and rates can be raised above
competitive levels. Proponents of the theory cite evidence that sizeable premium
increases in the past have been followed by periods of high profitability. They find
the data marshalled by the industry in support of current price increases inconclusive
and therefore self-serving, and they occasionally charge that recent price increases are
intended to recover losses suffered in previous years.
The weakness of the conspiracy theory is that it cannot explain how a market that
recently was characterized by intense price competition has turned anticompetitive.
During the four years prior to 1983, for example, commercial liability insurance was
widely available at level prices, notwithstanding the double digit inflation that
prevailed during the period. Maintaining a nationwide cartel in an industry as
unconcentrated as property/casualty insurance would be extremely difficult, as this
recent history suggests." Moreover, in a competitive market the recapture of past
supra note 2.
ORGANIZATION,
10. See NAIoAL IsNRA ECCONSUMER
D;r OFJusTnce, TORr PouicE
11. For discussion of the degree of concentration in the industry, see UNnErDSTATEs
Vomm GRouP, AN UPDATE
ON THELiAa'n Cists, app. 9-10 (March, 1987).
HeinOnline -- 48 Ohio St. L.J. 401 1987
OHIO STATE LAW JOURNAL
[Vol. 48:399
losses through current price increases is impossible, since past losses are in effect
sunk costs that do not burden potential new entrants.
Finally, the current crisis not only involves the affordability of coverage, but its
availability as well. Liability insurance for certain risks-directors and officers,
nurse-midwives, day-care centers, bars and restaurants, obstetricians practicing in
certain settings-was unavailable at any price for months at a time in some states
during 1986. As George Priest has suggested,' 2 it would be a strange cartel indeed
that exercised its monopoly power by refusing to sell its product to some of those who
wished to purchase it, even at very high prices. In short, the conspiracy explanation
for the liability insurance crisis is both theoretically implausible and empirically
unproven. 13
Despite the weakness of the conspiracy theory as a general explanation for the
crisis, I think it likely that some companies took advantage of the crisis and raised
premiums to levels higher than they would have raised them had there not been
widespread turmoil. Here I have in mind the lines sold in states where-because the
market is thin and the costs of entry are high-the one or two insurers who write
coverage have a niche that cannot be immediately undercut. For example, a medical
malpractice insurer which sells ninety percent of the coverage in a state with only two
thousand physicians may be able to double its premium rates for certain classes of
physicians without facing serious rate competition in the short run.14 In many states
where there is open price competition in most commercial lines, regulatory approval
of rate increases is nonetheless required for niche-market lines such as medical
malpractice. But the difficulty of predicting liability exposure makes assessing the
15
justification for a rate increase in such lines a highly speculative exercise.
Regulatory disapproval of proposed rate increases in such lines therefore is not as
common as might be expected.16
A final issue that fits most comfortably in a discussion of the conspiracy theory
is the insurance industry's role in promoting the tort reform legislation that has been
adopted by many states in response to the insurance crisis. Although it may seem
unsurprising to find the industry favoring limitations on tort liability, that stance in
fact proves more complicated to explain than a superficial look at the issue might
suggest. One would think that property/casualty insurers would favor the expansion
of tort liability, because the greater the threat of liability, the greater the demand for
liability insurance. Why might it be in the interest of insurers, then, to support tort
reforms that limit the incidence and scope of tort liability?
There are three plausible reasons, only one of which smacks of something like
conspiracy. First, the industry may recognize that there are practical limits on the
percentage of individual and corporate income that can be spent for insurance. When
supra note 7.
12. See G. PRInes,
ox LIABILrTINSURANCE,
supra note 1, at 9.
13. See I GovERNoR's COMMISSION
14. In the long run, competition may develop either from the commercial market or through the establishment of
medical mutuals.
15. See infra Part IV.
16. For a discussion of the problems faced by state insurance commissioners see K. ABR.Atv, Dm emtreso PsK:
INSURANc
E, LEGALTHEORY,AND PUBuCPoucY 38-41 (1986).
HeinOnline -- 48 Ohio St. L.J. 402 1987
1987]
MAKING SENSE OF THE LIABILITY CRISIS
insurance prices exceed these limits, responses that are not in the industry's interest
may be triggered. Risk retention and self-insurance alternatives that would not
otherwise be considered may come into play,1 7 more stringent regulation may be
proposed, and the insurance industry may be blamed for the escalation of the costs of
the tort system. Self-interest may therefore suggest that tort reform be favored,
notwithstanding that reform would limit demand for the industry's product. But this
is hardly conspiracy against the interests of the insuring public.
Second, even apart from the above considerations, all expansions of tort liability
are not necessarily in the insurance industry's long-term interest. An industry that
relies on the certainty afforded by the law of large numbers may favor slow, steady,
and predictable expansion of tort liability. But when the choice is between expansion
that creates exposure that is difficult to predict with precision, and limitations on tort
liability that facilitate predictability, the latter are likely to be preferable. Whatever
one concludes about the wisdom of this posture, its motive is not the desire to engage
in price gouging.
A third explanation for the industry's support of tort reform, however, is
consistent with the conspiracy theory. This view suggests that the industry simply has
shorter-term interests in mind in favoring reform. Having calculated premiums based
on the supposition that tort liabilities would remain steady or expand, the industry
then argued for reforms that will reduce its cost base and produce greater profits than
its earlier calculations would have anticipated. In the long-run this approach may also
reduce total revenue, but over the short-term it improves profitability. Moreover, if
tort liability again begins expanding some years after enactment of the reforms, then
revenue increases will have been delayed, but not denied.
In one sense this argument is irrefutable. The reduction in insurers' esposure
under policies already sold that would result from tort reform cannot be an entirely
unpleasant prospect for them. It is therefore not surprising that, in light of this
prospect, some state legislatures demanded rollbacks in premium increases as a
condition of the enactment of the 1986 reforms.1 8 In another sense, however, the
argument infers a conspiratorial motive from behavior that could just as plausibly be
construed as innocent conduct. If the 1985-1986 premium increases resulted from
what insurers perceive as an increasingly unpredictable expansion of tort liability,
then the industry's support of tort reform can be primarily ascribed to its desire that
this expansion be controlled. Any windfall gained through the reduction of exposure
under pre-existing policies should be seen as a mere side effect of reform, albeit a
welcome one. On the other hand, if the 1986 premium increases were the result of
anticompetitive price gouging, why would the industry then support legislative
reduction in the demand for its product, having already proved itself capable of taking
monopoly profits? Why kill the goose that laid the golden egg?
17. For discussion of the increasing use of alternatives to the commercial market, see BusinessInsurance, Jan. 26,
1987, at 14-16.
18. Florida is the most publicized example. For discussion of the legislation that produced this rollback, see
Schulte, Availability, Affordability, and Accountability: Regulatory Reform of Insurance, 14 FLA.ST. U.L. REv. 557
(1986).
HeinOnline -- 48 Ohio St. L.J. 403 1987
OHIO STATE LAW JOURNAL
[Vol. 48:399
The available answers are none too persuasive. Perhaps the industry's support
for tort reform was a colossal fraud, designed to deceive those who questioned the
premium increases that prompted legislative consideration of reform. According to
this view the industry supported reform, but did not actually favor it. Or perhaps the
industry's support for tort reform was simply another example of its supposed
preference for short-term over long-term advantage. Under this view the industry
chose to capture the gain from reduced liabilities under pre-existing policies rather
than continuing to maximize its monopoly profits. However, the persuasiveness of
this explanation depends not only on the assumption that it was a cartel which raised
premiums above competitive levels, but also that, having maintained the solidarity
necessary for the price increase, the cartel "members" then agreed through their
trade associations to favor their short-term rather than long-term interests by
supporting tort reform. A consensus on such a strategically complex choice would be
highly improbable in the absence of any indication of the attending debate finding its
way into the public record.
While these suppositions may be possible, they are speculative and highly
unlikely. Yet their implausibility depends at least in part on the credibility of the
alternative explanation for the industry's support for tort reform-that the growth of
enterprise liability has made the prediction of tort liability and the pricing of insurance
against the risk of that liability increasingly difficult. I now turn to that explanation.
IV.
ENTERPRISE LIABILITY AND THE PREDICTABILITY PROBLEM
Although the parameters of the growth in tort liability over the past decade have
not been definitively quantified, there is little doubt that it has been substantial. There
are two ways to measure the magnitude of this growth. The first method is to measure
increases in the cost of items that are components of tort recoveries. By this measure,
the increases have been sizeable. Between 1975 and 1984, for example, inflation was
eighty-three percent, real per capita income grew seventeen percent, and real health
care costs increased twenty-three percent. 19 Based on these factors alone, one would
expect an increase in the cost of liability insurance of approximately one hundred
percent during this period.
The second method of measuring increases is a direct comparison of payouts.
This measure also reflects a significant growth in liability. Since 1981, for example,
the average annual increase in amounts paid in non-automobile liability claims has
been seventeen percent, and the average annual increase in counsel fees for insured
defendants in such cases has been fifteen percent. 20 During the same period, the
average annual increase in the Consumer Price Index was seven percent. New York
City, which is self-insured for tort liability, experienced a four hundred percent
increase in amounts paid in tort settlements and judgments between 1978 and 1985.21
AWARDS:
HAVEJU.s RuN Waw? (Consumer Federation of America, May
TmNDS INLIABILITY
19. See M. CooPER,
1986).
ANDCOMPENSAnON
PAIDIN TORTLrmoAToN (Rand Institute for Civil Justice,
20. See J. KAKduK & N. PACE,COSTS
supra note 1, at 5.
ON LIABILITYINSURANCE,
ADVISORY
COMMISSION
21. See GOVERNOR'S
HeinOnline -- 48 Ohio St. L.J. 404 1987
1987]
MAKING SENSE OF THE LIABILITY CRISIS
In addition, between 1979 and 1984 the Gross National Product grew by about fifty
percent, while losses covered by commercial liability insurance increased one
22
hundred thirty percent.
It is no surprise that increases of this magnitude in tort costs would affect
liability insurance premium levels. What is surprising, however, is that the increases
were so sudden. A partial reason for the precipitous increases may lie in the
underwriting cycle. The cutthroat competition that resulted in "cash-flow" underwriting early in the decade may have delayed the impact of increased tort liability on
premiums for several years, until the forces of competition for market share finally
had to yield to more rational pricing. Nonetheless, the premium increases of 1985 and
1986 were so dramatic and, in some cases, so enormous, that prudence suggests a
search for other explanations as well.
The most plausible explanation for the size and suddenness of the premium
increases is a decline in the property/casualty insurance industry's confidence that it
could predict the scope of the liabilities it would face under the policies it sold after
1985. Thus, on top of the increases that can be ascribed to the operation of the
underwriting cycle and to predictable increases in tort liabilities, an additional
premium was assessed to cover insurers against the risk that their liabilities would
escalate in unanticipated ways. This "unpredicability risk premium" is probably
responsible for the remaining component of the recent round of increases.
There is little evidence to explain why insurers' confidence dropped so
precipitously in 1985. Surely the severe tightening of the reinsurance market at the
end of 1984 is part of the answer, although this only pushes the inquiry a step
backward. In any case, there may well have developed a kind of communal mind-set
among primary insurers that is not entirely inconsistent with some of the suppositions
of the conspiracy theory. But there is no way to determine whether this mind-set was
symptomatic of paranoia or of realism until the claims lodged against the holders of
policies issued during the past two years accumulate in the years to come. It is
possible, however, to pinpoint a number of legal developments that are likely to have
shaken insurers' confidence in their ability to predict future liabilities.
First, scientific advances have made it possible to connect exposure to hazardous
materials to an increased long-term probability of contracting certain diseases. The
result has been the proliferation of claims for long-latency diseases allegedly caused
by such exposures. Instead of rendering liability for such diseases more insurable,
however, these scientific advances often have produced the opposite effect. Experts
are sometimes able to retrospectively identify the cause of the claimant's disease with
sufficient certainty to support the imposition of liability. But science has not
progressed sufficiently to enable the accurate prediction of the probability and
severity of the occurrence of diseases that may be associated with a particular product
or material in order to correctly price insurance against liability for these diseases.
This problem is especially severe in attempting to calculate premiums for
"occurrence" coverage, which insures against liability arising out of the insured's
22. See IsURANcE Smvics Om,
INsuri
PRo'rrnarn:THE FActs 25-27 (1986).
HeinOnline -- 48 Ohio St. L.J. 405 1987
OHIO STATE LAW JOURNAL
[Vol. 48:399
activities during the policy period, regardless of the year when injuries allegedly
caused by these activities manifest themselves. In pricing occurrence coverage the
insurer must predict all liabilities that may eventually be incurred because of this
year's activities by the insured. When these liabilities include an uncertain quantity
of long-latency disease, making this prediction is especially difficult. An alternative
is to offer "claims-made" coverage, which insures against all claims first made
during the policy year, regardless of the year when the activity causing the claim
occurred. Because the claims-made insurer need predict only next year's claims in
setting a price for next year's coverage, its task is much simpler. But claims-made
coverage shifts much of the risk of long-term liability onto the insured, and thus
provides less real insurance protection than occurrence coverage. For this reason, the
industry encountered serious opposition to its effort to introduce a claims-made policy
form in commercial liability lines in 1986.
A second factor contributing to unpredictability is legal change. Although the
expansion of liability need not automatically undermine predictability, recent legal
changes have done so in several ways. Because many of the expansions of tort liability
that have occurred during the past decade were not anticipated by insurers, they have
become wary of their ability to predict future expansion. The role played by this kind
of wariness should be emphasized, even though it cannot be documented statistically
and sometimes seems unwarranted. Insurance underwriters have become highly
distrustful of courts and juries. This distrust can often obscure relevant distinctions
between states with narrow and those with liberal rules of recovery, and between
standards adopted by obscure trial courts and those endorsed at the appellate level.
There is little, aside from several years without major legal surprises, that is likely to
neutralize this wariness.
The kinds of legal decisions about which insurers are most concerned are those
that create new forms of exposure that are much less predictable than those they
supplement or displace. Several examples are worth noting, because they illustrate
the manner in which new forms of liability that sometimes seem logical from the
standpoint of the tort system can severely hinder insurability. Many of these standards
do not yet and may never represent mainstream principles of tort liability; but the
prospect that they may take hold cannot be ignored by insurers.
First, a few judicial decisions have imposed what amounts to retroactive strict
liability for product defects for the failure to wam users or consumers of products
about risks that were not and could not have been known at the time a warning could
have been given. 23 The difficulty of predicting liability for failing to discover or to
wam of risks that could not have been known at the time a product was made is
obvious. Pricing an occurrence policy under such circumstances is an exercise in
speculation. Pricing claims-made coverage is easier, since the injuries or diseases not
anticipated in the past will have begun to materialize when a price must be fixed. But
23. See, e.g., Dart v. Wiebe Mfg., Inc., 147 Ariz. 242, 709 P.2d 876 (1985); Elmore v. Owens-Illinois, Inc., 673
S.W.2d 434 (Mo. 1984); Beshada v. Johns-Manville Products Corp., 90 N.J. 191, 447 A.2d 539 (1982).
HeinOnline -- 48 Ohio St. L.J. 406 1987
1987]
MAKING SENSE OF THE LIABILITY CRISIS
because the insured activity has already occurred when a claims-made approach is
used, it may be troubled by adverse selection.
Second, some courts have expanded the scope of joint and several liability
among independent tortfeasors, either directly or by shifting burdens of proof. 24 The
Federal "Superfund" statute25 has a similar effect in creating liability for the costs of
cleaning up hazardous waste disposal sites. 26 Predicting the magnitude of a particular
insured's potential joint and several liability is particularly difficult for two reasons:
whether an insured will be liable at all and how much damage will occur hinge partly
on the behavior of other parties, and the portion of any judgment an insured must pay
depends on the assets available from the other defendants. Yet these other parties
often will be unidentifiable at the time an insurance pricing decision is made.
Third, courts in several jurisdictions have begun to allow awards of
noneconomic damages to individuals exposed to danger before they suffer any
physical harm. Thus far these awards have taken the form of medical monitoring
expenses. 2 7 But plaintiffs are now beginning to claim compensation-even apart
from these losses-for the mere exposure to risk.28 The class of individuals who
either are exposed to risk and may therefore need medical monitoring, or fear that
they will contract disease, is much larger and less determinate than those who actually
do suffer physical harm from a given course of conduct. Consequently, predicting the
scope of liability that may be imposed for these new kinds of claims is considerably
more difficult than predicting the scope of more traditional forms of liability.
Fourth, there may be a move to incorporate into tort law the very broad standards
used to assess liability for the costs of cleaning up hazardous waste sites under the
Superfund Act. Personal injury actions brought in the vicinity of Superfund sites at
Times Beach, Love Canal, and Woburn, Massachusetts suggest the prospect of this
development, as do others. 2 9 Superfund liability is generally categorized as retroactive,
strict, and joint and several;30 a move to adopt these cleanup liability standards in tort
would cause new insurability problems for several of the reasons already discussed.
But the Superfund liability analogy, if fully implemented, would add another
difficulty as well. Under the Superfund Act, liability for clean-up costs is imposed not
only on the present and past owners or operators of hazardous waste disposal sites,
but also on the parties whose industrial processes generated the waste and on the
24. See, e.g., Borel v. Fibreboard Paper Prod. Corp., 493 F.2d 1076 (5th Cir. 1973), cert. denied, 419 U.S. 869
(1974).
25. 42 U.S.C. §§ 9601-9615 (1982 & Supp. 111 1985).
26. See United States v. Chem-Dyne Corp., 572 F. Supp. 802, 810 (S.D. Ohio 1983).
27. See, e.g., Friends For All Children, Inc. v. Lockheed Aircraft Corp., 87 F.R.D. 560 (D.D.C. 1980) (Court
granted preliminary injunction ordering defendant to provide interim medical treatment). In a subsequent proceeding, the
court ordered the defendant to provide diagnostic examinations but refused to order the defendant to pay for interim
medical treatment. Friends For All Children, Inc. v. Lockheed Aircraft Corp., 587 F. Supp. 180, 184-86 (D.C.), aff'd,
746 F.2d 816, 831 (D.C. Cir. 1984).
28. See, e.g., Payton v. Abbott Laboratories, 386 Mass. 540, 437 N.E.2d 171 (1982); Ayers v. Township of
Jackson, 189 N.J. Super. 561, 461 A.2d 184 (1983). For a proposal that compensation be awarded on the basis of risk,
see Robinson, ProbabilisticCausation andCompensation for Tortious Risk, 14 J. LrGA. SwD. 779 (1985).
29. See, e.g., Kenney v. Scientific, Inc., 204 N.J. Super. 228, 497 A.2d 1310 (1985).
30. See 42 U.S.C. § 9607 (1982).
HeinOnline -- 48 Ohio St. L.J. 407 1987
OHIO STATE LAW JOURNAL
[Vol. 48:399
parties who transported it to disposal sites. 3 1 These are liabilities unknown at
common law. More importantly, if the generators and transporters of toxic waste are
subjected to retroactive, strict, and joint and several liability for personal injuries
caused by exposure to material leaking from sites in which these parties once
deposited waste, calculation of their liability insurance premiums will be fraught with
risk. Their insurers will have to assess not only the riskiness of their own insureds'
conduct, but also the volume, toxicity, tendency to migrate, and durability of the
waste deposited in a site by other parties, for these factors will influence the extent
of any given insured's potential tort liability should the site leak. Such liability is very
likely to be uninsurable.
Each of the foregoing developments has expanded the liability of insured
enterprises in a manner that makes predicting the extent of their potential liability
more difficult. A very different form of legal change, however, also has undermined
predictability. In. a variety of settings courts have interpreted liability insurance
policies to afford insurance against events that insurers had not anticipated covering.
For example, judicial interpretations of policy provisions governing whether there
was an "occurrence" within the meaning of a policy, 32 whether cleanup costs are
covered by a Comprehensive General Liability policy, 3 3 the applicability of the
"pollution exclusion,' '34 the number of insured "occurrences" resulting from events
taking place over a period of years, 35 and the insured "events" triggering coverage
against liability for long-latency diseases 36 all have produced coverage obligations
which insurers had not anticipated.
Insurers can react to novel judicial interpretations by either raising the price of
future coverage provided under identically phrased policies, or by changing the
language of these policies to clarify their coverage obligations. Unfortunately,
however, neither of these reactions is satisfactory protection if the courts continue to
adopt readings which, in the insurers' view, rewrite policy provisions rather than
interpret them. The climate of uncertainty surrounding the drafting and interpretation
of insurance policies is thus another factor that contributes to the unpredictability that
influenced the dramatic escalation of insurance premiums that occurred in 1985 and
1986.
Each of the foregoing developments lends support to the unpredictability
hypothesis. Unfortunately, the very nature of the liability insurance business makes
verification or refutation of the hypothesis impossible without psychoanalyzing those
responsible for the actuarial calculations that precede rate-setting. Only the future will
tell us whether recent increases have been excessive or whether, on the contrary, the
31. 42 U.S.C. § 9607(a) (1982).
32. See Buckeye Union Ins. Co. v. Liberty Solvents and Chem. Co., 17 Ohio App. 3d 127, 131-32, 477 N.E. 2d
1227, 1233 (1984).
33. See Chemical Applications Co., Inc. v. Home Indem. Co., 425 F. Supp. 777 (D. Mass. 1977).
34. See Jackson Township Mun. Utils. Auth. v. Hartford Accident & Indem. Co., 186 N.J. Super. 156, 164-66,
451 A.2d 990, 994-95 (Law Div. 1982).
35. See Jackson Township Mun. Utils. Auth. v. American Home Ins. Co., No. L-29236-8 (N.J. Super. Court, Law
Div. 1984) (appeal pending).
36. See Keene Corp. v. Ins. Co. of North America, 667 F.2d 1034 (D.C. cir. 1981), cert. denied, 455 U.S. 1007
(1982).
HeinOnline -- 48 Ohio St. L.J. 408 1987
1987]
MAKING SENSE OF THE LIABILITY CRISIS
industry's apparent fears about the explosion of tort liability reflect the prescience for
which the industry strives.
V. INCENTIVES TO LITIGATE
The fourth theory I shall examine takes a comparative approach. This theory
suggests that the incentives to litigate personal injury cases in the United States are
so strong that the long-term trend will ineluctably be toward more tort liability and
greater liability insurance costs. This theory does not contradict the other three; it
merely denigrates their importance. The theory does not purport to explain the
dynamics of the recent liability insurance crisis. Rather, it sees recent developments
as blips on a graph that traces a steadily upward course toward higher costs.
The theory points to three features of the American tort system that create
distinctive incentives to litigate: the availability of contingent fees for plaintiffs'
attorneys, the "American rule" that a losing party is not obligated to pay his
adversary's costs, and the absence of the widespread, generous forms of social
insurance prevalent in other Western democracies. The contingent fee system and the
American rule make the instigation of a tort claim much less risky for the American
plaintiff, because he risks so much less than potential plaintiffs in other systems. In
effect, the plaintiff sells a portion of his claim to his attorney in return for counsel fees,
and he does not risk liability for his opponent's fees. The absence of generous sources
of social insurance also creates incentives to sue for losses not otherwise compensated,
and knowledge of the probable absence of full first-party insurance protection by the
victim makes juries more sympathetic to plaintiffs' claims than they would be under
other circumstances.
The consequence of these incentives, according to the theory, is two-fold. Not
only are more tort claims filed and higher judgments awarded, but plaintiffs and their
attorneys also are continually challenging the limits of liability, and the occassional
success in pushing back those limits (either legal or monetary) sets a precedent,
creates publicity, and encourages more litigation. This self-reinforcing cycle provides
the impetus behind the increased costs of liability and liability insurance.
The weakness of this theory, of course, is that it is too general to be of any use
in understanding what has happened to the liability insurance market in the past year
or two. On the other hand, its strength is that it points to systemwide characteristics
that may well have more influence over the long-term than any of the more "local"
considerations I have canvassed thus far. The theory is a sober warning against
quick-fix solutions: superficial doctrinal modifications, renewed regulatory attention
to insurance rates, controls on contingent fee levels. Such reforms may rachet down
costs temporarily, but costs will continue to rise if the theory is correct. The theory
suggests that even application of the antitrust laws to the insurance industry,
regulatory attempts to mitigate the underwriting cycle, and retrenchments in modern
tort liability doctrine are unlikely to eliminate the fundamental cause of the insurance
crisis. To obtain substantial savings in the cost of liability insurance, the theory
argues, there will have to be substantial reduction in the incentives our system
provides injured parties to litigate their claims.
HeinOnline -- 48 Ohio St. L.J. 409 1987
OHIO STATE LAW JOURNAL
[Vol. 48:399
VI. CONCLUSION
Each of the four theories I have examined has elements of plausibility. Indeed,
I accept them all, with the caveats that I have noted. Standing alone, however, no
single theory is capable of explaining the causes behind the liability insurance crisis
that has recently struck the country. Taken together, the accurate elements of each
begin to provide a satisfactory explanation for the causes of the crisis.
Fashioning a cure for the crisis is, unfortunately, more difficult than explaining
it. More focused regulatory attention on the thin and niche markets where competitive
forces do not adequately restrain premium rates would help neutralize monopolistic
behavior where it does exist and the underwriting cycle will inevitably progress on its
own. Therefore, the availability crisis will probably abate, as both underwriting
capacity increases and new entrants into markets where coverage has been unavailable offer it, albeit at higher prices. For example, the trade publications have recently
noted that nurse-midwives and day-care centers are locating sources of coverage that
was simply unavailable during much of 1986. Other tight markets should loosen up
in a similar fashion.
But these are short-term solutions. The longer-term problems of unpredictability
and litigation incentives pose fundamental issues that are not so easily resolved.
While insurance markets are probably more flexible and resilient than many critics of
modem tort liability suggest, these markets possess neither infinite resilience nor
infinite flexibility. Ultimately there is a tension between the predictability that
produces stability in insurance markets and the risk-reducing, cost spreading goals
that undergird so much of modem tort liability. If there is to be stability in the
insurance markets that are so important to our economic well-being, we may have to
place some limits on the scope of tort liabilities. Without such limits, uncertainty may
continue to operate like a broad-based, undiscriminating liability insurance tax that
simply drives up costs across-the-board.
How is this "uncertainty" tax to be eliminated? In his General Theory, John
Maynard Keynes suggested that "[p]ractical men, who believe themselves to be quite
exempt from any intellectual influences, are usually the slaves of some defunct
economist.' '37 It may be that the present generation of jurists and juries is the slave
of outdated legal theories, and that only a generational shift can bring a change in the
legal climate. In any event, the tendency of judges and juries to treat defendants'
insurance coverage like social insurance for victims is unlikely to be neutralized by
tort reform alone. In the long-term, doctrinal reform probably cannot successfully
solve the tort system's problems until alternative sources of compensation are
available as substitutes for tort recoveries. Otherwise, reforms likely will follow the
defense of those who must implement them through individual lawsuits; judges and
juries simply will manage to circumvent the new limits on liability and reinstate the
old system.
Until new, or at least broadened forms of compensation outside the tort system
are developed, the compensation/entitlement mentality is likely to dominate personal
37. See J.
KEYNEs,THE GENERAL
THEORY
OF EMp.owyr,
Itnywr ANDMoNEy 383 (1936).
HeinOnline -- 48 Ohio St. L.J. 410 1987
19871
MAKING SENSE OF THE LIABILITY CRISIS
411
injury litigation. The "new compensation" that is necessary might take a number of
different forms-catastrophic health insurance, expanded private disability insurance, special-purpose administrative compensation funds, or more generous taxfinanced social security protection. Discussion of the strengths and weaknesses of
these different approaches is beyond the scope of this article; but the liability
insurance problem is unlikely to abate until these two issues-compensation and
insurability-are recognized as opposite sides of the same problem.
HeinOnline -- 48 Ohio St. L.J. 411 1987
HeinOnline -- 48 Ohio St. L.J. 412 1987